Experimental Evaluation of Staff Ability and Performance Based on Depression Level Analysis Using Machine Learning Principle

With regard to multifaceted aspects of life, depression has huge impacts running into millions of people all over the world and is associated with performance at work. The paper compares a suitability assessment regarding diverse machine learning models predicting kind of influence depression may pose onto the performance by an employee. This is done specifically using the Ensembled DTANN model. Here, it turns out to be a Kaggle collection dataset with 4,532 entries having demographic, job-specific, and psychological assessment data. Through intense preprocessing of the data and feature selection, the Ensembled DTANN model was developed by integrating the concept of Decision Trees and Artificial Neural Networks. The comparison analysis depicts the supremacy of the Ensembled DTANN model because it gained maximum accuracy of 93.5%. Precision, recall, and F1 score metrics similarly produced high scores, thus evidence that the model is resilient in dealing with complexities of mental health data. It hence supports the capacity of hybrid models to effectively quantify nuances of mental health impacts on occupational performance.